Abstract
Background: Generally, the efforts to predict antidepressant use from patient demographic factors have not been fruitful.
Aim: Our objective was to generate hypotheses regarding antidepressant use among older primary care patients.
Methods: We utilized a mixed methods design that is both hypothesis-testing and hypothesis-generating. Adults aged 65 years and over were recruited from primary care practices and interviewed in their homes. We examined the personal characteristics of older adults according to antidepressant use (hypothesis-testing). Participants taking antidepressants and participants not taking antidepressants were asked open-ended questions about their views on treatment for depression. Themes related to use of antidepressants were examined (hypothesis-generating).
Results: Older adults taking antidepressants were more likely to be white and have more depression symptoms compared to older adults not taking antidepressants (p < 0.001 and p = 0.004, respectively). Positive and negative themes emerged when participants discussed antidepressant use. We linked quantitative data from the participants with the themes they endorsed to form an emerging theory about older adults' perceptions about antidepressant use.
Conclusion: Few personal characteristics were associated with antidepressant use. An improved understanding of how older adults view antidepressant use, derived from multiple methods, may inform clinical practice.